Epithelial ovarian cancer (EOC) encompasses several subtypes of tumors that exhibit distinct clinicopathologic features. Our overall goals are to delineate the molecular pathways that regulate the histogenesis of different subtypes of EOCs, and to determine the role of aberrant differentiation in transformation of ovarian surface epithelial (OSE) cells. This endeavor is a critical step to defining novel molecular markers and targets that enable development of more effective multiplexed approaches for EOC diagnosis and "designer" therapeutics that target specific types of EOCs. EOCs are thought to arise from the simple epithelium lining the ovarian surface. EOCs differ from many other types of epithelial tumors in that their differentiation patterns are often more complex than that of the precursor cell. Indeed, the major subtypes of EOCs are characterized by their architectural resemblance to the specialized epithelia of the reproductive tract that derive from the mullerian ducts. HOX genes control growth and determine the unique identity of each tissue during normal development. Studies of mouse mullerian duct development indicate that the Abd B-like HOX genes, hoxa9, hoxa10 and hoxa11, normally regulate morphogenesis of the fallopian tubes, uterus and cervix, respectively. In preliminary studies, we analyzed epithelial and stromal expression patterns of the human Abd B-like proteins in microarrays of EOCs, and found striking parallels between their tumor subtype-specificity and their distribution in the normal reproductive tract. Our preliminary studies also revealed roles for HOXB7 and HOXA7 in promoting aberrant proliferation and differentiation, respectively, of OSE cells, and for HOXA9 and HOXA7 in neoplastic transformation. We hypothesize that transformation of OSE cells and the morphogenesis of the major subtypes of EOCs arise through aberrant expression of HOXA7, HOXB7 and the Abd B-like HOX genes. In this proposal, we will determine the roles of these HOX genes in (1) differentiation of tumorigenic and non-tumorigenic ovarian epithelial cells along specific mullerian-like pathways, (2) neoplastic transformation of OSE cells, and (3) epithelial-stromal interactions in EOCs. These studies therefore investigate two fundamental but poorly understood aspects of EOC histogenesis, namely, the morphologic heterogeneity of the disease, and the relationship between aberrant differentiation of OSE cells and neoplastic transformation.